Machine That Changed The World

The lean production concept has become popular in America industries after publish the book “Machine That Changed The World” being published by Womack in 1991. That book helps the reader to understand the data and motivation behind the lean movement. However, the book does not go into detail on the lean implementation process. This limitation has been covered by another publication from Womack and Jones in 1996 with the title “Lean Thinking”. This book shows how to create an industrial world in which workers share the challenges and satisfactions of the business.


The concept of lean production is a process of value creation. It begins from defining value for a specific product or service, to creating the value through value stream, eliminating or minimizing the interruption along the value stream, pulling value and finally pursuing perfection. The value was sustained by the characteristic lean culture, such as

  • Team – based work organization with multi – skilled workers who are capable of taking high degree of responsibility for work within their areas;
  • Active shop floor problem – solving and continuous improvement,
  • Maintaining low inventory, quality management by prevention rather than detection,
  • High commitment human resource policies which encourage a sense of shared destiny within a factory,
  • Maintain smaller supplier group and share or exchange of business information with suppliers,
  • Encouraging of cross – functional development teams
  • Retailing and distribution channels which provide close links to the customer and permit a make – to – order strategy to operate.

Many researchers have shown that company that adopted lean production system reported significant improvement in overall productivity that lead to enhanced company financial performance. The case study carried out by Michael A. Lewis (2000) shows that, the companies under his study have shown significant in operation improvement after transforming from conventional batch production system to lean production system.

Process Engineering Data Analysis

Receive a call from process engineering manager, she request me to arrange statistical analysis training for her process engineers. She wants the engineers to have better understanding of statistical analysis and apply it in their daily work.

I am willing to customize training for my colleague, but I don’t know what to deliver. I think of this problem for the past few days, nothing finalized.

Tonight, I have a chance to really sit down to review my old training material and summarize something which is useful for them.

Step 1 : Understand your process
All process deliver result, result is the outcome of combination of various process input variable. The relationship between input and output could be translate into mathematical term as below:

Any changes of input variable will resultant of output change.

If you do not understand well the relationship of input and output variable, you may end up collecting data which is not correlated.

Step 2 : Establish problem statement

Click here to link to previous post talk about problem statement establishment

Step 3 : Objective setting
Objective only can be set full understanding of problem statement.

If you don’t know your problem well, you don’t what to focus, and you may have problem to set a right objective.

Once a clear objective set, you will know the information should be collect for statistical analysis.

Step 4 : Establish data collection plan
Plan your data collection schedule or interval. In most cases, data collection for manufacturing process should be carry out in a stable process condition.

Anyway, there are cases where process condition was purposely tweak into unstable condition for data collection purpose.

It it depend on the objective of the data collection.
Once you have the data collection. Here come to the part of statistical analysis

Step 5 : Summarize your data
Most statistical analysis software (such as Minitab) provide feature of generate data summary report.

Information can be obtain from the summary report are
Mean / median
Data distribution and spread
Normality test
Confident interval
Standard deviation
And others

The report provide a rough idea of your data behavior.

Step 6 : Check for equal variance
Check for equal variance is recommended if data was collected from various sub group, the purpose of the assessment is to understand the difference between each data set.

Box plot is the common graphical method I like to use for quick assessment.

I may apply Barttlett’s test or Levene’s test if I would like to have proper statistical assessment.

Step 7 : Process stability check
Main purpose of process stability assessment is to understand the process condition when data was being collected.

Time serious, I-MR and X – Bar R chart are the most commonly use graphical presentation in most process engineering.

The assessments provide an idea of process behavior over time.

Step 8 : Process capability analysis
Process capability analysis tells a process capability to produce output that meeting predefined target.

Click here to link to previous post talk about process capability analysis for normal data
Click here to link to previous post talk about process capability analysis for non normal data

Step 9 : Conclude your finding
Conclude your finding based on information obtain from Step 5 – 8

ANOVA Example (Continue From Last Post)

Step 1 : Normality Test


Step 3 : Process stability assessment


Step 4 : Process capability assessment


Step 5 : Test for equal variance


Step 6 : ANOVA


Conclusion :

  1. ANOVA analysis shows no significant difference of mean between M/C1 M/C2 and M/C3. However data of M/C 1 and M/C 3 show mean at the lower side to the control dimension.
  2. Test for equal variance shows data from 3 machine having equal variance, but M/C1 produce highest variation among the 3.
  3. Process capability analysis show Cpk value at 0.79. The process in general not capable.
  4. Process capability analysis show process is stable and no special cause happen in the process.
  5. Normality test show all data collected are normal data

Advice to my colleague :

Quite risky to mix the parts into one bin due to concern of machine to machine variation.

He need to do something to minimize machine to machine variation, bring the process mean to the target

He can consider to mix the parts once the line stabilized.

.

ANOVA Example

A colleague show me some data that he collect from 3 different machines and ask for help to perform some statistical analysis.

He would like to assess the consistency of the part dimension that produce from 3 different machines, to make decision whether to mix into 1 bin or pack it separately according to machine number.

I am trying to use ANOVA for the data analysis

The sample data given to me as below

Control dimension = 15.000

M/C1

M/C2

M/C3

13.548

13.655

14.469

16.353

13.793

15.409

13.456

15.263

14.006

14.246

16.279

13.935

15.091

14.716

13.264

15.329

15.770

13.986

12.926

15.154

13.206

12.705

14.339

13.972

14.581

15.204

15.913

13.111

15.611

15.731

Below are the steps I perform the data analysis

Step 1: Normality Test

Minitab procedure: Stat >> Basic Statistic >> Normality test

Note: ANOVA only can apply on Normal Data


Step 2: Stack up the data into single column

Minitab procedure: Data >> Stack >> Columns

Step 3: Check for process stability

Minitab procedure: Stat >> Control Charts >> Variable Charts For Individuals >> I-MR

Step 4: Check for process capability

Minitab procedure: Stat >> Quality Tool >> Capability Analysis >> Normal


Step 5: Test for equal variance

Minitab procedure: Stat >> ANOVA >> Test for equal variances

Note: To assess whether the machine running under same std dev condition or not.


Step 6: ANOVA

Minitab procedure: Stat >> ANOVA >> One Way

Done….

I will discuss the analysis result on my next post…


Quality Control With 300% Visual Inspection

Receive an 8D report from my colleague sometime ago regarding a process related issue. One of the proposed containment actions that shocked me is the implementation of 300% visual inspection right after the problematic process station.

I guess he knew it is not a good solution, due to pressure of delivery commitment, this is the best he can do in this kind of unstable process condition. At least he can sort out some good part and ship it to the customer.

My question is, is 300% visual inspection more effective than 100% inspection?

I m not sure.

My experience tell me, if something that cannot be screen out through 100% visual inspection, that particular defect will exist even the same batch of parts were screen through again and again.

Why?

This is because the property of the defect is not obvious enough to spot by the visual inspection operators. If the defect slips through the first inspection, it will slip through the next inspection and so on…

So, what is the point having so many round of inspection? Overly inspection doesn’t ensure defective parts were totally remove from the batch, whereas it induce others problem such as accuracy between operators and others defect due to excessive handling.

Do you ever face the same problem? You have process capability issue at one hand, customer delivery commitment on the other hand and your boss standing behind you with an angry face.

What should you do?
You are welcome to leave comment on this case.

Six Sigma Family

Have a chat with my MBB few days ago, we chat about Six Sigma of course.

He brought up quite an interesting topic. He says that most Six Sigma web sites today only focus on Six Sigma application in business process improvement. Application of Six Sigma in solving family problem like husband and wife relationship, kids education, shopping, personal financial issues were rarely touch by the Six Sigma practitioners.

Yes… his idea makes sense. Six Sigma DMAIC is a very powerful problem solving tool. If it can be use to solve multi million business cases, why can’t it use to solve family problem?

I can’t find an excuse to say “No”

Hmmm… I should think about this topic and do something different with Six Sigma DMAIC.

Anyway, some re-customization is needed to make it more user friendly. Something have to replace the boring statistical analysis.

Well, let me have sometime to think about it, will update this topic under label Six Sigma Family.

Attitude Problem Again

Every Six Sigma deployment involved money, attitude and time.

Money
Money is required for Six Sigma knowledge development and infrastructure set up. In general, usually money is not an issue for any deployment since management aware and budget allocated.

Attitude
My personal opinion, people’s attitude is the main factors that determine the program sustainability. Many Six Sigma program die off due to lack of mental support.

Question: What kind of attitude I expect?

Answer
The attitude to believe continuous improvement is the only way to stay ahead

The attitude to believe every single improvement they did will help to improve overall organizational efficiency.

The attitude to believe problem or challenges can be overcome in a systematic manner; DMAIC is one of them.

Time
It is very much depend on the environment where Six Sigma was plant. Time to deliver result in a Theory Y environment would be much shorter than Theory X environment.

Higher people’s resistance to change mean longer time required to achieve the target. Bottom line, we go back to people’s attitude problem again.

Black Belt Hunting

Received several phone calls from head hunting agents, there are hunting Six Sigma Black Belt for their client, mostly rich US based MNC located at Penang, KL and JB. There are several local companies in the list as well. I did not consider their offer mainly due to the work location; it is too far from my home. My top priority now is to train up my two little GB.


Anyway, glad to know that Six Sigma program is getting popularized in local industries, the demand of BB is increasing. Good, at least I know I m on the right track.

More and more companies are adopting the program, hopefully a healthy growth of DMAIC practice can be seen in the local industries.